Driven and Passive Trends

In his work, McShea also considers the possible causes that have been proposed to explain evolutionary trends, and highlights a broad distinction between driven and passive accounts (e.g. [McShea 94], [McShea 98]). A driven trend can occur when a directed, pervasive force exists in a system, causing a particular measure to move in one direction rather than another. However, McShea points out that trends can occur in measures even in the absence of such driving forces (i.e. passive trends). These may occur when some sort of boundary exists in the system, representing a minimum (or maximum) below (above) which the measure is unable to move. The distinguishing feature of such boundaries, compared to the biasing forces that cause driven trends, is that they only operate on a small subset of the state space. As a simple (if somewhat implausible) example, imagine a multicellular organism (consisting of a very small number of cells) which is free to evolve both in the direction of more cells and in the direction of fewer cells, with no selection pressure favouring evolution one way or the other. Both larger and smaller organisms arise by chance as evolution proceeds, so the diversity of observed sizes increases over time. Now, imagine that there is no particular upper limit on the size of the organisms which may evolve (at least, perhaps, until they consist of millions of cells). However, there must be a lower boundary on organism size--each organism must comprise at least one cell. In this situation, the mean observed organism size will increase, due to the existence of a boundary but in the absence of any directed force.

McShea suggests various ways by which these two causes may be distinguished, but remarks that ``most explanations that have been proposed for complexity trends implicitly invoke biases and thus are driven ... In contrast, little has been said about possible causes of boundaries, a subject which is ripe for deeper theoretical investigation'' [McShea 96] (p.486). In our work on artificial evolutionary systems it is therefore important to remember that some of the patterns of life observed in the biological world may not be due to pervasive directed forces, but rather to the existence of boundaries.